This proposal describes a series of laboratory experiments designed to ascertain the conditions under which victims of a crime of petty larceny are likely to report the incident to the police. The experiments involve systematic manipulation of several variables presumed to affect victim crime reporting. The variables relate to characteristics of (a) the crime and its setting, (b) the thief, and (c) the victim. Among the specific aspects of the crime which are being studied are (1) the magnitude of the theft, (2) the advice given by co-victims and bystanders, and (3) whether the thief is still on the premises at the time of reporting. Characteristics of the thief which are being studied include the thief's (1) race, (2) sex, and (3) socioeconomic status. Victim characteristics which are being studied relate to (1) personality, (2) demographic, and (3) attitudinal factors. The research also investigates the accuracy of victims' immediate and delayed recall of the theft and their reactions to it. The information thus obtained will provide data bearing on the accuracy of victimization surveys as they relate to the crime of larceny.